The proximal ⅓ of femur is one of the most common locations for benign bony lesions. Bone tumors such as giant cell tumors, chondroblastomas, aneurysmal bone cysts, fibrous dysplasias, and simple bone cysts often are seen in this region. Clinical symptoms include pain, limp, and leg-length inequality. Nonoperative measures are the first choice of treatment. Occasionally, the marked bony destruction renders this area at risk for a pathologic fracture. These tumors usually are benign, grow slowly, and are anatomically deep to surrounding soft tissues. Symptomatic involvement or pathologic fractures of the proximal part of the femur frequently require surgical intervention to achieve pain relief and resumption of unrestricted activity. Reconstruction is difficult because of the size of the lesion and resulting bony defect.
The authors report the results of 35 patients who had intralesional removal of tumor followed by allogeneic cortical strut grafts and autogenous cancellous bone grafting in conjunction with internal fixation.
MATERIALS AND METHODS
Seventy-three patients with benign bone lesions of the femoral neck and trochanter were surgically treated at the authors' institution from 1988 to 1991. Small lesions that can be treated with simple bone grafting, pinning, or palliative methods were excluded from this series. A small lesion was defined as a volume defect of approximately 60 cc, dimensions smaller than 2.5 cm, and with a thick cortex of adjacent bone present. En bloc resection and hip arthroplasty provided the best chance for cure initially for patients in whom the lesion was so extensive that the articular surface could not be preserved. These patients are not included in this study.
Thirty-eight patients treated with combined curettage, combination bone grafting using autogenous and allogeneic materials, and internal fixation were seen in followup. Three patients were lost to followup. The average duration of followup was 3 years 6 months, (range, 2-5 years). The most common symptoms were pain and limp. Eleven patients had pathologic fractures. The indications for operation were large cystic lesions with impending or actual fractures and a progressively enlarged defect that could not be treated with simple bone grafting procedures. Three criteria for impending pathologic fractures are: (1) 2.5 cm or larger lytic lesion in the upper femur, (2) a lesion involving half of the femoral neck or more, and (3) expansive lesion with thin cortex and progressive deformity. All patients had a definitive pathologic diagnosis. The surgical procedure for all patients consisted of intralesional removal of the tumor and replacement by a deep frozen allogeneic cortical strut graft. A sliding screw and autogenous cancellous bone graft were used in conjunction with the strut graft. No spica cast was needed.
Under an image intensifier, the patient was placed in a lateral position on the operative table. Either the anterolateral or posterolateral approach over the trochanter and neck region was used, depending on the location of the lesion. The leg was rotated, and the soft tissue was divided to expose the cortex of the neck and trochanter area. Through a cortical window, a representative biopsy specimen was obtained for frozen section. The preferred position of the cortical window was extraarticularly. The definitive surgical procedure was done after the diagnosis of the frozen section was established. If the preoperative diagnosis did not correspond with the intraoperative diagnosis, a staged operation was done. Otherwise, the lesion was approached directly, and a wide window was made in the cortex to allow adequate tumor removal and completion of the bone stabilization procedure. The margin was scraped with a curette under direct vision after exteriorization and tumor removal. A dental burr was used to extend the margins beyond the reactive zone of the tumors. Local adjuvant treatment with phenol and alcohol were used to extend the margin, especially when dealing with giant cell tumors and fibrous dysplasia.
The volume of defect after tumor removal in 22 cases was 93 cc (range, 55-150 cc). The wide exteriorization cortex allowed easy application of the cylinder cortical strut and internal fixation device. If indicated, the fracture was reduced, and a guide K wire was inserted through the trochanter to the neck and head region. The position and alignment were checked with an image intensifier. The lateral cortex and femoral head were reamed under the direction of a guide wire. The preferred position of the lag screw is inferior to the femoral neck and head. Fluoroscopic image intensifier determined the correct position and alignment.
A cylinder cortical strut was prepared from banked ulnar, radial, or fibular bone. The struts were preserved with the deep frozen (-70 °C) method. One or more struts were used over the defect in the neck, trochanteric, and subtrochanteric regions if the space defect was large enough. A cylinder cortical strut graft a little longer than the defect extending from head to lateral cortex of trochanter was obtained. The graft was inserted into the defect and through the lateral cortex by pulling it into the femoral head proximally and pushing it into the left lateral cortex of the greater trochanter laterally. The cortical strut was placed in an adequate position and fixed with temporary wire especially for pathologic fracture cases. A hip compression screw was inserted after the deep frozen allogeneic cortical strut was fixed. The preferred position of the large lag screw is inferior to the cortical strut to prevent the screw from cutting through. In addition, using the mechanical properties of the cortical strut keep the length of the femoral neck and prevent collapse of the femoral head and neck. The autogenous cancellous bone graft that was taken from the posterior iliac crest was used to fill the remaining defect and was fixed with a side plate (Fig 1).
The patients were allowed to ambulate with partial weightbearing during the initial postoperative period and to be fully weightbearing after 6 weeks. No cast was required. Patients were seen in followup every 3 months for the first year after surgery and then at 6-month intervals. For security, the internal fixation device was not removed for an average of 3 years.
The overall results were evaluated using the Musculoskeletal Tumor Society's rating score of limb salvage for the proximal thigh, hip, and pelvic girdle.3 The seven primary factors considered were (1) motion, (2) pain, (3) stability, (4) deformity, (5) strength, (6) functional activity, and (7) emotional acceptance. Each of these primary factors was rated as excellent, good, fair, or poor. In the excellent category, 6 of the 7 primary factors must rate excellent. The seventh may be good, fair, or poor. In the good rating, 6 of the 7 primary factors must rate good or better. The seventh may rate fair or poor. For a rating of fair, 6 of 7 primary factors must rate fair or better, and the seventh may rate poor. In a poor rating, 2 or more of the primary factors must rate poor.
Of the 35 patients, 16 were women and 19 were men. The average age of the patients was 27 years (range, 18-54 years). The pathologic diagnoses were 8 aneurysmal bone cysts (Fig 2); 14 monostotic fibrous dysplasias (Fig 3); 11 simple bone cysts (Fig 4); and 2 giant cell tumors.
Good bony incorporation was seen on the radiographs of all patients by 3 and 6 months after surgery. Remodeling of the implanted allograft strut and good healing was observed in radiographs at 6 and 9 months after surgery. There were no complications, including infections or fractures of the allograft bone, femur, or metallic implant. Local recurrences and avascular necrosis of the femoral head were not found. All patients healed well and returned to full weightbearing activities and normal leg function. The overall results using the Musculoskeletal Tumor Society grading system were excellent.
The upper third of the femur is 1 of the most common sites of benign tumor involvement. The specific anatomic location makes the management of these lesions unique. Most of the lesions are small and silent. They need close observation or palliative treatment, especially in children. Occasionally, the lesions are large, with progressive growth, fracture, and deformity. Surgical procedures are required. Treatment should eradicate the lesion sufficiently to preclude recurrence while restoring bone integrity to prevent a subsequent pathologic fracture or deformity. Reconstruction of large bone defects presents a major challenge in orthopaedic practice. There are various methods for filling the cavity with bone grafts or other substitutes.7,9
The most important factor for effective curettage is adequate exposure. Curettage through a limited cortical window has a relatively high reported incidence of local recurrence. The method requires a large cortical window and complete excision of the tumor. After tumor removal, the margins must be scraped thoroughly with a sharp curette, followed by the use of a power dental burr to extend the margins beyond the reactive zone of the tumor. Local adjuvant treatment with phenol may be applied to extend the margin.2,12 Phenol is a cytotoxic substance that coagulates protein. Phenol is effective in reaching the crevices in the cortical bone to which the tumor has permeated and killing any surface residue within the tumor bed.
In recent studies, autografts have proven superior to allografts in the rate and completeness of healing for solitary large lesions. An autogenous bone graft using the fibula or rib has been suggested as the preferred method for dealing with large defects.1,7 Autogenous cancellous grafts have rapid revascularization and act as effective substitutions. It may be impossible to obtain sufficient quantities of autogenous graft in younger patients, making an additional operation necessary. It is difficult to get an autogenous cancellous bone graft greater than 60 cc, and the defect of the lesion often is greater than this size.5 Vascularized iliac bone graft transfer to bridge the defect in the proximal femur requires additional surgical time and microsurgical technique.9
Allograft cancellous bone can be absorbed, and eventually an osteolytic lesion can develop that needs biopsy to determine if the defect has recurred. In the current series, there were no recurrences. Deep frozen allografts undergo biologic changes that are temporally inferior to an autograft. Allografts have the advantage of an unlimited supply and no additional donor site morbidity.13 The process of incorporation of allografts is slower and probably less complete than that associated with autografts.1,6 Deep frozen allografts are associated with decreased immunogenicity and show no demonstrable change in initial mechanical properties.
Bone cement has no biologic capacity, and it is hard to predict whether and how long the plugged bone can tolerate full weightbearing without mechanical failure, especially around the femoral neck area. Thus, bone cement is not the preferred choice for filling a large defect in this location.11 Biocompatible materials offer the surgeon another option, but they have not achieved complete acceptance.8
The authors' preferred surgical technique for benign or aggressive, large benign bone tumors is a combination of intralesional excision, phenol cauterization, allogenic cortical strut, and autogenous cancellous bone grafting. The framework of the femoral neck and trochanter become weak after a wide window is made and the tumor is removed. The defect needs reconstruction to restore hip joint function. The authors use this surgical plan because the deep frozen allograft cortical strut in experimental animal models and humans is associated with decreased immunogenicity4 and has shown no demonstrable change in initial mechanical properties.10 The sequence, but not the timing, of these events makes the process of incorporation of allografts the same as that used for fresh autografts. As the graft is revascularized, remodeling begins, which presumably enhances incorporation of the graft. The defect initially is filled with a lag screw and cortical strut. The remaining space is limited. The amount of cancellous bone graft required easily can be taken from the iliac crest. The internal fixation device allows good support during bone healing that limits the risk of a pathologic fracture and the need for a plaster cast.
The procedures in this study are an alternative to, and do not necessarily replace, more conventional surgical methods. There are 2 advantages to these procedures compared with simple autogenous cancellous bone grafting: (1) They provide enough biomaterial to fill a defect that is greater than 60 cc. One or more cortical struts can be used to fill the defect over the neck, trochanteric, or subtrochanteric areas after subtotal resection of the tumor (Fig 3); and (2) The cortical struts act as a semistructure graft, providing mechanical support and avoiding collapse of the fracture of the neck after dynamic hip screwing during the early healing stage. Simple cancellous bone grafting cannot achieve this purpose (Figs 2,4).
The authors use this method for preserving vascularity, allowing rapid incorporation of the allograft, and providing good mechanical support. It uses biologically safe and active materials. The subtotal excision can be completed with rapid healing of the defect and with minimal complications. This unique reconstruction provides increased strength and prevents deformity or fractures in the upper third of the femur. The patient may return to normal activity without functional impairment.
1. Burchardt H: The biology of bone graft repair. Clin Orthop 174:28-42, 1983.
2. Capanna R, Fabbri N, Bettelli G: Curettage of giant cell tumor of bone: The effect of surgical technique and adjuvants on local recurrence rate. Chir Organi Mov 75(Suppl 1):206, 1990.
3. Enneking WF: Modification of the System for the Functional Evaluation of Surgical Management of Musculoskeletal Tumors. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 626-639, 1987.
4. Friedlaender GE: Immune responses to osteochondral allograft. Current knowledge and future directions. Clin Orthop 174:58-68, 1983.
5. Glancy GL, Brugioni DJ, Eilert RE, Chang FM: Autograft versus allograft for benign lesions in children. Clin Orthop 262:28-33, 1991.
6. Heiple KG, Chase SW, Herndon CH: A comparative study of the healing process following different types of bone transplantation. J Bone Joint Surg 45A:1593-1616, 1963.
7. Jaffe KA, Dunham WK: Treatment of benign lesions of the femoral head and neck. Clin Orthop 257:134-137, 1990.
8. Jarcho M: Calcium phosphate ceramics as hard tissue prosthetics. Clin Orthop 157:259-278, 1981.
9. Leung PC, Chow YYN: Reconstruction of proximal femoral defects with a vascular-pedicled graft. J Bone Joint Surg 66B:32-37, 1984.
10. Pelker RR, Friedlaender GE, Markham TC, Panjabi MM, Moen CJ: Effects of freezing and freeze-drying on the biomechanical properties of rat bone. J Orthop Res 1:405-411, 1984.
11. Persson BM, Wouters HW: Curettage and acrylic cementation in surgery of giant cell tumors of bone. Clin Orthop 120:125-133, 1979.
12. Rock M: Adjuvant management of benign tumors: Basic concepts of phenol and cement use. Chir Organi Mov 75:195-198, 1989.
13. Thompson RC, Pickvance EA, Garry D: Fractures in large-segment allografts. J Bone Joint Surg 75A:1663-1673, 1993.